TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM Paper • 2607.27205 • Published 3 days ago • 122
Characterizing Warp Divergence from Pascal to Blackwell Paper • 2607.23402 • Published 6 days ago • 6
Read It Back: Pretrained MLLMs Are Zero-Shot Reward Models for Text-to-Image Generation Paper • 2607.11886 • Published 19 days ago • 84
AgenticDataBench: A Comprehensive Benchmark for Data Agents Paper • 2607.01647 • Published 30 days ago • 37
Not All Disagreement Is Learnable: Token Teachability in On-Policy Distillation Paper • 2605.26844 • Published May 26 • 26
Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding Paper • 2605.29707 • Published May 28 • 152